Download - Portland State University


Oct 24, 2013 (4 years and 8 months ago)



Two phase flow, flow of both a liquid and gas state, is crucial in
countless space applications including power cycles, the storage and
transfer of liquid fuels, cryogens, propellants, life support systems,
and thermal control systems.

However, truly steady and fully
developed flow has not yet been
achieved in low
g experiments due
to limitations in hardware, size,
cost, and time.

Presently, necessary applications
have been either over
designed or
designed to completely avoid two
phase flow conditions.


Previous experimentation has been overwhelmingly dominated by
pressure driven flow. Large complex systems have been used to
force a liquid slug or a gas bubble through a long straight tube.



To Syringe








These systems can be large, difficult to control, expensive, complex,
and unable to achieve steady
state conditions.

Den Chen, 1986

Fluid Behavior

As a liquid slug passes through a tube, a thin film is deposited along the
inside walls. This film thickness fluctuates depending upon the
gravitational conditions. If correctly predicted and modeled, system
parameters such as pressure drop and heat transfer rate can be calculated
simplifying designs by reducing unnecessary size, weight, and cost.

In reduced gravity, typically
neglected forces become more
dominant and may even
control the flow regime. In the
case of two
phase slug flow,
surface tension becomes a
prevailing force and must be
included in all analysis.

More Fluid Behavior

In a reduced gravity environment, it is difficult to assure of the location
of liquid and gas phases in a two
phase system. Similar to a 1
environment, an applied external force can be exploited to collect and
position the liquid into a single continuous slug.

An unusual idea of applying a simple
centrifugal force on a closed circular
loop became the new model for this
experiment. This unique solution
eliminates the control concerns of
pressure driven flow, reduces the
size, weight and complexity of the
experiment, and will ultimately allow
for a steady
state flow field.

1g Design

The simplicity of the reduced gravity design allows for testing and
modeling in a 1g environment. Steady flow is easily achieved by
mounting a tube of fluid on a circular disk. To further predict fluid
behavior in reduced gravity, the disk is tilted at various angles,
therefore lessening the effect of gravity in the direction of flow.

In addition, various tube sizes,
fluid viscosities, and rotational
speeds are examined to
determine the most appropriate
combination for the one chance
reduced gravity experiment.

1g Results

For verification purposes, the 1g data was compared with several
published predictions. The closed circuit loop demonstrated that
state slug flow is easily achieved and well
predicted by
established analyses.

A New Solution

A small disk holding several sealed tubes of a liquid and gas mixture is
mounted above a larger plate. By rotating the large plate, a centrifugal
force is applied to the small disk positioning one continuous slug to the
outside rim. The second smaller disk can then be rotated resulting in
phase slug flow.

This simple design allows for
fine control of both rotational
speeds while quickly providing
state, fully
slug flow.

Reduced Gravity Design

The experiment is enclosed in an aluminum frame and undergoes a
rigorous safety evaluation to NASA specifications. Due to the
unusual environment aboard the reduced gravity aircraft, individual
components must be designed to withstand severe loadings, as high
as nine times gravity in certain directions.


A remote camera above the
small disk records and
transmits data to an onboard
video display. While the flow
is monitored during the flight,
the images are analyzed at a
later date. The video is
downloaded and the flow
properties scrutinized using
image analysis software.

From these images, film
thicknesses, dynamic contact
angles, deposition rates and
transient and steady state flow
regimes are examined.


Balakotaiah, Vemuri and Larry Witte. “Studies on Two
Phase Flows at Normal and

Microgravity Conditions.” 1998/1999. Institute for Space Systems and Operations.

29 June 1999 <

Chen, Jing
Den. “Measuring the Film Thickness Surrounding a Bubble Inside a Capillary.”

Journal of Colloid and Interface Science.

109 (1986): 341

Colin, C., J.Fabre, and A.E. Dukler. “Gas
Liquid Flow at Microgravity Conditions.”

Int. J. Multiphase Flow.

17 (1991): 533

Hallinan, Dr. Kevin, and Jeffery S. Allen and Jack Lekan. “Capillary
Driven Heat Transfer

(CHT) Investigation, MSL
1, STS
83.” 2002. NASA. August, 2002.


Incropera, Frank P., and David P. Dewitt.
Fundamentals of Heat and Mass Transfer

New York: John Wiley & Sons, 2001: 357.

Taylor, G.I. “Deposition of a viscous fluid on the wall of a tube.”
Journal of Fluid


10 (1961): 161

Young, Donald F., and Bruce R. Munson and Theodore H. Okiishi.
A Brief Introduction to

Fluid Mechanics.

New York: John Wiley & Sons, 2001: 304

Special Thanks
: The NASA Reduced Gravity Student Opportunity Program,
the Oregon Space Grant Program, Mark Weislogel, Portland State University’s
College of Engineering and Computer Science, Infinity Images, FMC Allen
Machinery, and Chehalem Machine Works